Hearing in Transitional Mammals: Predictions from the Middle-Ear Anatomy and Hearing Capabilities of Extant Mammals

The structure and evolution of the mandible, suspensorium, and stapes of mammal-like reptiles and early mammals are examined in an attempt to determine how, why, and when in phylogeny the precursors of the mammalian tympanic bone, malleus, and incus (postdentary jaw elements and quadrate) came to function in the reception of air-borne sound. The following conclusions are reached: It is possible that at no stage in mammalian phylogeny was there a middle ear similar to that of "typical" living reptiles, with a postquadrate tympanic membrane contracted by an extrastapes. The aquamosal sulcus of cynodonts and other therapsids, usually thought to have housed a long external acoustic meatus, possibly held a depressor mandibulae muscle. In therapsids an air-filled chamber (recessus mandibularis of Westoll) extended deep to the reflected lamina and into the depression (external fossa) on the outer aspect of the angular element. A similar chamber was present in sphenacodontids but pterygoideus musculature occupied the small external fossa. The thin tissues superficial to the recessus mandibularis served as eardrum. Primitively, vibrations reached the stapes mainly via the anterior hyoid cornu, but in dicynodonts, therocephalians, and cynodants vibrations passed mainly or exclusively from mandible to quadrate to stapes and the reflected lamina was a component of the eardrum. In the therapsid phase of mammalian phylogeny, auditory adaptation was an important aspect of jaw evolution. Auditory efficiency, and sensitivity to higher sound frequencies were enhanced by diminution and loosening of the postdentary elements and quadrate, along with transference of musculature from postdentary elements to the dentary. These changes were made possible by associated modifications, including posterior expansion of the dentary. Establishment of a dentary-squamosal articulation permitted continuation of these trends, leading to the definitive mammalian condition, with no major change in auditory mechanism except that in most mammals (not monotremes) the angular, as tympanic, eventually bcame a non-vibrating structure.

The number of turns in the cochlear spiral and length of the basilar membrane in several mammalian species were compared with the octave range and the high-and low-frequency limits of hearing. Basilar membrane length and the number of spiral turns were not related. Among ground dwelling mammals, the number of turns in the cochlear spiral was more strongly related to octave range than was basilar membrane length. Basilar membrane length was inversely related to the high-and low-frequency limits of hearing. The best estimates of high-and low-frequency limits and octave range were derived from formulas which included both the number of turns in the cochlear spiral and the basilar membrane length as factors. The number of turns in the cochlear spiral was most highly correlated with the difference between the low-frequency limit of hearing and the lowest frequency mechanically analyzed by the traveling-wave envelope, peak-shift property of the basilar membrane [von Békésy, Experiments in Hearing (McGraw-Hill, New York, 1960)]. The coefficient of correlation for the number of spiral turns and the octave difference between the lowest audible frequency and the lowest frequency distributed as a unique point of maximum displacement along the basilar membrane was r = 0.997 (P less than .001) at 60 dB SPL. Mechanisms by which the spiral form of the cochlea may affect the motion of hair cells and the selective response of the tectorial membrane to differences among traveling-wave envelope slopes and peak locations were reviewed. It was proposed that in ground dwelling mammals, the spiral form of the cochlea extends the octave range of hearing and that through mechanisms such as these increases the sensitivity of the cochlea to frequencies below the low-frequency peak-shift limit of the basilar membrane.